Predisposition to develop melanoma results from a complex, multifactorial etiology that involves genetic factors, endogenous physiology, and exogenous exposures. The interaction of these factors to cause melanoma may differ by age, race, or family history. Knowledge about multiple risk factors in melanoma etiology will aid in the identification of individuals at increased risk for developing melanoma. Ultimately, this knowledge may facilitate melanoma prevention strategies. We propose to study candidate genetic biomarkers of melanoma risk, and determine their association with melanoma phenotypes. The underlying biological hypothesis of the proposed research is that susceptibility to develop melanoma stems from the complex interactions of multiple factors, including inherited genotypes associated with immune surveillance. To address this hypothesis, we will use a multidisciplinary approach that involves methods drawn from molecular biology and epidemiology to evaluate the role of genes involved in melanogenesis and development of a melanoma. We propose to consider the role candidate melanogenesis genes (i.e., cytokines and other immune surveillance genes (ISG)), the interaction of these genes and other melanoma risk factors (e.g., skin/hair/eye pigmentation, previous UV exposure, history of dysplastic nevi, and family history of melanoma) in the etiology of melanoma. Using the existing infrastructure of the Pigmented Lesion Group at the University of Pennsylvania, we propose a case-control study that will directly address the complex, multifactorial etiology of melanoma that involves the interaction of genotypes involved in immune surveillance. This study will address a number of specific hypotheses. First, we will characterize the frequency distribution of alleles at candidate susceptibility genotypes. Second, we will evaluate whether genotypes are associated with tumor characteristics including stage, grade, and age at diagnosis. Finally, we will evaluate whether these genotypes are involved in melanoma etiology, and whether these genotypes interact with one another and other melanoma risk factors. In order to address these hypotheses, we will undertake a study using the extensive resources of the Pigmented Lesion Group at the University of Pennsylvania. The sample will consist of 1000 melanoma cases and 1000 controls. Risk factor information will be obtained by questionnaire, a DNA biosample will be collected using a non-invasive cheek swab method, and diagnostic pathology information will be collected using a systematic approach. Analyses will be undertaken to evaluate the role of candidate genotypes and other risk factors in melanoma etiology, including genotype by environment interactions.